Mariama Ndoye – Modelling energy consumption patterns in residential buildings

This thesis is set against the backdrop of the energy transition. Indeed, environmental issues such as air pollution, global warming, and the depletion of fossil fuels are at the heart of major contemporary debates. The energy transition is therefore necessary to mitigate the effects of energy consumption, whose actual impacts are devastating on a global scale. Within this framework, the building sector emerges as a major driver of energy consumption and greenhouse gas emissions. This research aims to model the energy consumption patterns of residential buildings following a renovation project. The study of energy behavior in residential buildings remains both poorly understood and underestimated in models. This challenge stems from two obstacles: a lack of usable data on actual household energy consumption; and the complexity involved in conducting surveys on households’ domestic energy behavior. The goal is to study, analyze, and dynamically model the behavior of residential buildings, taking into account the interactions between space, households, and energy consumption. The study will focus on the impact of the dynamics of household energy behavior on the building’s final energy consumption. A survey will be conducted before and after renovation to collect sociodemographic data on households, household characteristics, and, most importantly, household behaviors. The data from these surveys, along with data from sensors installed as part of the André project, will be analyzed using various statistical models.